Emule Future Development
Posted 21 January 2016 - 08:02 PM
Don't tell me to code. I'm not interested in development of another eMule for Windows platform. My days of P2P development is over, been there done that. This thread is NOT about adopting eMule and emulating Torrent as we can keep the core principle without compromising and still move forward into the future.
Posted 22 January 2016 - 06:29 AM
And there is the first problem. Even David, Tuxman and Wizard have so different opinions about the future of emule. So which of them has the right opinion, which is wrong.
And even about the topic "what is emule" and "what should emule become" are so different sights in the community.
Be p2p, be more like torrent, focus to an healthy network, focus an speed.
Posted 23 January 2016 - 04:23 PM
Now let say one selected developer wants to add something and he has support of the community and other developers agrees with him then Some Support shouldn't be the final voice to stop it. At the same time he doesn't have to step down as he can monitor and guide them for improvement.
You can make improvement without compromising the core value of eMule's principles. Yes all MODers has their own ideas but i'm sure they can all agree on something and go from there by putting it to vote. It's stupid to have 50 MODS while the main is behind. I mean lets be honest here we all know the main ED2K/KAD client now is not eMule, as its chink VeryCD leeching mod. I have checked stats on my own end from 6 nodes which so far interacted with over 90k users and 1% is mldonkey, 3% is lphant, 19% aMule and rest 77% is emule. But out of that 57% is you guys guessed it single VeryCD mod and rest is variant of eMule and its mods. How the hell does that happen?
Posted 31 January 2016 - 10:03 AM
No development = no eMule news = continuation user base shrinking.
This post has been edited by nemoW: 31 January 2016 - 10:07 AM
Posted 31 January 2016 - 10:05 AM
At the end we need someone that is able to provide a good management of this: users can submit feedback, developer are clearly guided where to focus and they agree on the focus. It is not a matter of 'X please step back', sure an official change would help but in the short period only.
Extended signature: click.
Posted 31 January 2016 - 11:33 AM
I agree with you there nemoW as its general concept which seem to elude Some Support for past DECADE (yes and I do mean decade) as little to no growth only pushing users away. That is not a claim that’s actual fact as shrinking userbase year per year is evident.
pier4r if it was so easy why wasn’t it done before? The whole idea is to push official base upto date and not drop ANOTHER mod client. Which will only be targeted to pure uploaders or leechers but very ineffective against the mass. As it would be the ONLY way to move the current userbase as it was done during KAD's implementation.
I mean come on a decade later still no NAT-T or IPv6 implemented, WTF? While half of current ED2K userbase is LowID. Where LowID2LowID can be implemented, current DHT half ass KAD can be revised, CS, QR can be revised as well as we all know the problem since last decade as its been discussed so many times and STILL nothing.
Posted 31 January 2016 - 03:37 PM
I've seen P2P clients that demanded to be the latest possible version, otherwise you couldn't boot into the P2P network. It wasn't a physical limitation, it was a measure that forces users to run latest version. Indirectly, it becomes a physical limitation because keeping clients up to date at all times allows you to evolve a client WAY faster. Small changes that make huge difference. And if something goes wrong, it's also easy to backtrack since you're doing 1 change at a time and not 500 changes at once. Which will allow developers to experiment easier, debug easier, induce new features and changes faster and evolve the client in shorter time with less manpower.
I may not be a programmer so bash me for that aspect all you want, but I've been a tester of so many programs I know how things operate. And when an otherwise active program doesn't receive a single update in 6 years timeframe, you know things are far from "OK". eMule needs to evolve faster, become more user friendly in terms of booting into the network, more automated to get rid of LowID etc.
Posted 01 February 2016 - 12:21 AM
So yes, other developers should take over. But in my opinion, that's what forks are for. It's basically how eMule started and it should be how a newer, more active client succeeds it. Why not hand over the website just to someone? Because the official client might be inactive, but it still guarantees the same ideals with which it earned the trust in the first place. No malware, building the foundation for a working network, no glaring bugs, no ads / spyware, open source and so on. This isn't something given, much smaller open source projects have run into those problems (like shareaza years ago where the domain was sold and used for adware).
And there do exists some real forks / new clients already. But none I'd like to see being published on this website. One of the first things David's client did for example was going closed source (and I think it still is, but I haven't checked it for a while). And they concentrated included torrent/filellockers. It's a different vision for a client and that's fine. But users should decide to switch to that, rather than having eMule replaced with such a fork.
Due to the fact that eMule always had a closed developer team (which had its advantages and made a lot of what it became possible) there is no trusted developer structure in place to hand over the responsibility (which I'll admit I'm very hesitant to do in the first place for the reasons listed above). Plus one has to consider that there are less developers in general interested working on filesharing clients as the golden years when they were developed and pioneered are behind them and today it's more "boring" maintanance. Yes there are still a few active developers here too, like Fox who posts helpful code snippets but it's far away from, the manpower which would be needed to manage the whole project and possible revamp the client.
I do admit that the situation is far from perfect. But there is no easy solution like "hand it over to someone and everything will work out".
Posted 01 February 2016 - 08:12 AM
It's "boring maintenance" because nothing new is there to work with. If you have an old car with crappy little engine to tune, you'd get bored fast as well. But if you have a modern 4 cylinder block with tons of stuff to fiddle with, you'd be all thrilled, fiddling it all day long, day after day. That's what eMule was for modders when new builds with big new changes arrived. I just prefer car analogies, I like them because they work. And that's the problem with eMule. It's an old car with a little engine. Sure it works, is reliable because hardly anything can break down anyway, but it's old, outdated and you can't really fiddle with it anymore so you leave it in the garage.
I appreciate what original devs have done, but it's really time to move on. Change the policy by forcing the updating of client. Once you have that, you can experiment all you want. Even if you break something network wide, you can just force an old build again and all will be restored opposed to current method of restraining yourself constantly not to go out of line of the old rules. That's why eMule hasn't changed much, because of constant fear of breaking compatibility with old clients. And with that, everything turns into snail pace.
Posted 01 February 2016 - 08:43 AM
This premise is already wrong. The eMule protcol is written in a way that new extensions and enchantments can be added without breaking any compability. After all, that's how it works for the official version too, there was never any version which was incompatible with any older version (only exception is deprecating Kad1 but that was rather for the fact that it had been replaced for years). The only thing which stops forks/mod being able to do all they want is that they need to reach a certain threshold of users for new features to work well.
Posted 01 February 2016 - 09:20 AM
more releases with little changes that users see that the emule dev is activ
thats all what i wish.
some little gui updates for example
some automatic stuff or little fixes
every update is a little advertising for the emule network
This post has been edited by hooligan3000: 01 February 2016 - 09:21 AM
Posted 03 February 2016 - 05:15 AM
Lugdunum did a very fine job creating the ED2k eServer project as I myself deployed the v17.15 linux NPTL edition of it on 2 servers before they were shutdown by force half decade ago by you know who. As hosting them inside USA are not very viable and they are not very forgiving and open about it when operating it inside their backyard. But Lugdunum has his reasons for making the project proprietary to limit and hinder fake ED2K servers being operated by Media Corp's affiliates in absence of source code. But I do remember few years back there was 2nd eServer project alternative to Lugdunum named Satan eDonkey, which was reversed engineered from Lugdunum's current eServer tree. I never ran it as it was for Windows OS but few people here has tested it and the source code for it was sent to few eMule Dev's. As I remember Some Support even got a copy since people didn't trust it as it came in a time when virtually all ED2K eServers were being shutdown. ED2K has its flaws as its not perfect but the biggest flaw with 2G-P2P is also its greatest strength as compared to our 3G-P2P mock-up version of DHT known KAD.
Yes I do agree ~Mule need to do more advertising and should be more open on media networks like facebook/twitter as other network has for more exposure. We should think of creating 3 sections for it. As 1st being the Development Team, 2nd being the Support Team on forum/irc, and 3rd being Media/News Team posting update regarding software commits and news content. As you can see other networks like Torrent which already been doing it, we don't even have to look at other network just look at inside our own network as you will see with VeryCD community which is rich and active and has surpassed offical ED2K/KAD client eMule in terms of usage now. We should also think about operating unified content distribution for legal content that are open under free domain to be hosted exclusively on ED2K.
Posted 04 February 2016 - 07:15 AM
Posted 04 February 2016 - 01:45 PM
I agree with you there in general. But as we seen other developers has already made endless mods which is why its very saturated and in log term hinders actual growth of main project since not unified under single body, as in this case more options is not better. Its good thing they have done things and took the initiative upon themselves to change things since Some Support refuse to do anything about it. I also agree with you as the standard of eMule should be fully revised since we still using obsolete structure designed for last decade as its not up to date today.
With all due respect man you have drowned us all with your snail pace race now to standstill. You did make valid point regarding trust and faith in brand. As I'm aware of Shareaza troubles as I was there along with few others who modified aMule fork for eDonkey-BlackNET under Shareaza's standard. You need to accept few facts as I don't think even your in denial as you see eMule is dying slowly. As your the current leader of the project thus these responsibilities falls on you and it doesn't mean you have to step down. All you need to do is direct others for help with main project but you advise others to fork another extension is not helpful.
I mean the variant of mods has its place. But the main ED2K/KAD client on the network is eMule for Windows OS, where aMule for Mac/Linux and for hybrid network mldonkey supports all OS. MODers can make all the changes they want but they wouldn't be able to penetrate the userbase effetely for it to fully work. Only way it will work successfully is if its fully backed up by the community as you see VeryCD has done. You can have the fastest car in the world but when your stuck in traffic nowhere to move back or go forward on top of that on road thats already narrow, your in gridlock. So you need to stop hindering the projects future by babbling about increase in network overhead and system resource usage, as its moot point. No one runs eMule on other platform or different architecture they are not stupid they would just deploy aMule or mldonkey for headless.
Major changes will be very tedious but you can take minor steps by modifying the GUI where user is alerted they obtained LowID. Then they should be redirected where they are informed what LowID is and why its harmful and how to seek help to obtain HighID via forum/irc. For those who can't ever get HighID since unable to access hardware NAT-T should be implemented. IPv6 support also since alot of ISP now deploying it and for the rest just add LowID2LowID already. Fact that you think LowID2LowID will encourage people to do nothing since lazy is stupid. As you need to practice common sense in basic standard for project's threshold.
Posted 04 February 2016 - 05:17 PM
VeryCD is the good example of what I don't want to see happen with eMule. And it was never backed by a community, but by money.
But back to the point. I don't see much to discuss here so I'll just clarify my position: The official website (including this forum) will stay under our control and will - for now - publish the official eMule version, even if it stops being developed.
I do encourage other developers to go on from there. If there is a viable fork / mod /new client which has an active and responsible development we might forward the users to it, possible as far as promoting it in the version update notification within eMule. A new developer/team will have their own website and forum and as users switch, this can become the new "official" home of the (next) client.
Posted 04 February 2016 - 05:55 PM
"under our control"
Can you give me the current roster of those figures? developers, mods, contributors? You have not outlined once again anything regarding the future of eMule development goals. Who said to hand over website/forum control? You guys can still monitor it but be more open to future development in general. As you know and I know P2P is not as it use to be as Torrent became game changer and made other client's footprint extinct. They have the funding and they have the developers compared to us. So whats the edge there? How is prompting another fork going to do anything? the problem still exist as you choose not to deal with it.
Posted 04 February 2016 - 08:43 PM
what the.... the point is the network need the offi client with updates.
i like the offi dev rules...no bad features etc
but its time for example ipv6, user friendly gui an other way to help to fix port problemes etc.
i think it will help someone...
im really interested in your plan for the future
This post has been edited by hooligan3000: 04 February 2016 - 08:50 PM
Posted 05 February 2016 - 07:55 AM
Its only my uneducated guess but I think its psychologically very hard to change once mind if one was stuck to a bad idea for to long. Kind of like when otherwise rational people can not give up religion or similar as they wasted so much life on it that admitting even to one self that it was BS is psychologically to traumatic.
Saving overhead is eMules religion but in the days of 100+ mbit lines it is as useful as the kosher/halal limitations of some Abrahamic religions are in the day of refrigerators and modern medicine.
a.k.a. not at all sticking to it is like shooting one self in the foot on purpose, just dumb.
About an other dev developing a new client or fork, why should someone do that for a client that is already fading into obscurity? The only chance eMule has in my opinion on a Renaissance is for a multi network client that supports torrents to also support ed2k.
But than I would guess that will not be a "viable fork / mod /new client which has a responsible development", right?
NeoLoader is a new file sharing client, supporting ed2k/eMule, Bittorent and one click hosters,
it is the first client to be able to download form multiple networks the same file.
NL provides the first fully decentralized scalable torrent and DDL keyword search,
it implements an own novel anonymous file sharing network, providing anonymity and deniability to its users,
as well as many other new features.
It is written in C++ with Qt and is available for Windows, Linux and MacOS.